At the end of the noughties augmented reality is rather fashionable, but it wasn’t always thus. To understand the current trajectory of AR, it’s not the history of computing itself that we should turn to but the history of the future of computing. Today AR is where it’s at but in the recent past it was just a staging post on the way to two alternative futures: virtual reality on the one hand and ‘everyware’ on the other.
In the 1980s the future of reality was virtual, not augmented. According to cyberpunk novels the computer was something we would ‘jack-in’ to and interaction would be wholly ‘immersive’. In the early 1990s our virtual reality future seemed assured. An influential book by Howard Rheingold and the 1992 film The Lawnmower Man http://www.imdb.com/title/tt0104692/ made the real world seem dull by comparison. At around the same time, a radically different view of the future was emerging. The promise of ‘ubiquitous computing’ (or to use a less prosaic name for it: ‘everyware’) was that computers would be all pervasive – embedded into the environment. We wouldn’t have to immerse ourselves in computed reality; we wouldn’t even be aware of computers’ existence. With everyware computing, computers would subtly do whatever we need them to do whenever and wherever we need them to do it. The distinction between VR and everyware is like two views of paradise. For some theologians, paradise is distinct from and wholly unlike the world, and that’s what makes it heavenly. Paradise for others is a matter of selection – keeping the bits you like and removing the toil and unpleasantness that stops us enjoying them. Everyware is like a team of angels dedicated to making your life richer and easier by anticipating your needs and removing the drudge. VR is like being an angel. What would happen to us (computer users) in the virtual reality future? This turned out to be as complex and controversial a question as debates through the centuries about whether angels have internal organs. Virtual reality was conceived as a strange form of dualism. In VR the ‘world’ is no longer material, but it requires mediation by a very material combination of body and machine. So, in the future, our bodies would be co-opted for navigating the computed reality we would inhabit mentally. This new role for our bodies and senses – mediating ‘ideas’ – would make a change from our more traditional dualist conception of existence in which our minds guide us through the material world we inhabit physically. (The prospect of sex in VR was the subject of prurient speculation every bit as intense as the pious debates about the existence of angels’ sexual organs.) The ontology of everyware is simpler. In ubiquitous computing there is no computed reality at all – just reality. The humanity of computer users is unchallenged by everyware. AR on the other hand weaves multiple realities together with all the affordances of VR and everyware but without the philosophical limitations of either. In the early days, AR was a poor relation to both futures. It was seen either as intermediate step on the road to fully immersive virtual reality and as a rather clumsy attempt at ubiquitous computing. So what happened? When did augmenting reality seem like such a good idea?Actually, the most important stimulus to the recent interest in AR is the availability of essential technologies for AR, but that’s the subject of a future post. The recent change in the ontological status of AR comes from realising 3 things: 1. The real world is pretty cool after all, and a lot more fun than VR. Even top of the range data-gloves can’t compete with 4 billion years of evolution and a head-mounted display hides a lot more than it reveals. 2. Computed realities, it turns out, are a lot of fun too. We don’t always need to be shielded from them as long as they are built with users in mind. (Our computed environment is a bit like our built environment. In both cases it helps to put people at the heart of the process and recognise that it is not a pure engineering problem we are dealing with.) 3. The interaction between computed reality and the real world is the really interesting bit of computing. It is OK to mix them up thoroughly – not just gently layer one on top of the other.
The shift in perception about the world and about computed realities transforms AR from being merely a stopping off point on the path to the future of human/computer interaction to being right at the heart of the future. VR and everyware still have a place in the future of HCI, but unlike AR, neither is the future. (We could at this point return to the paradise metaphor and ask what does this new view of finely woven AR tell us about heaven and Earth? But let’s not go there.) As I hinted, the most exciting prospect for me is not just adding a layer of data to our experience of the world but thoroughly mixing them up – weaving computed realities in and out of social realities and physical realities. This goes beyond ‘mere’ augmented reality. At the very least we should be thinking not just about augmenting reality but also about how the real-world can be used to augment data. (I have discussed ‘reality augmented data’ or RAD elsewhere: RAD: Augmented reality turned on its head) Afterword: There are lots of problems with this account. A good reviewer would ask, for instance, ‘who saw AR as a clumsy attempt at ubiquitous computing?’ but it’s a blog post not a learned essay – just a bit of fun. I hope you don’t mind a few leaks because it surely aint water-tight.